1. Technical Field
Embodiments of the disclosure relate generally to the field of scaled enlargement of microscopic surface features and more particularly to a method and apparatus for creating a multiple parallax visual sensor by copying and scaling of a compound insect eye lens.
2. Description of the Related Art
In their current state, robotic systems are unable to accomplish fully automated tasks in unstructured environments. No general purpose robot can navigate through obstructed or cluttered environments, pickup and manipulate objects, assemble components, repair broken systems or manipulate other systems with a high degree of precision with no touch labor. Despite the gains in structured environments such as pick and place machines, precision of a similar nature is not possible in a unstructured or unregistered environment with unregistered objects.
The lack of these kinds of capability in current robotics means that this functionality cannot be transferred to military applications in the form of robotic or artificial intelligence assistance to combat or logistics personnel. The unique problem associated with such devices is that robotic systems do not have a simplified method of collecting three dimensional situational information and quickly correlating it for use in solving mission problems. The ability to overcome this issue may allow, but is not limited to, precision close proximity operations between space and surface vehicles, robotic assembly of systems in unstructured environments to the degree required for productive space operations, autonomous vehicles capable of duplicating the manipulative abilities of humans and autonomous systems which can duplicate the navigation and mobility displayed by most animals.
Other methods of determining 3 D situational awareness include the use of single cameras with motion and image processing algorithms, the use of plenoptic cameras, and dual camera for single parallax. The first of these is the method used by pick and place machines. Current systems require very structured environments with pre registration of parts on perforated tape spools or similar approaches. Plenoptic cameras generate limited parallax and are not currently widely used in the desired types of applications since extensive algorithmic support is required. Dual cameras providing single parallax do not provide the level of capability that simplicity, functionality and reliability require for fully autonomous operations or geometric understanding of the position of a vehicle or device relative to a known or unknown environment or other device. Determination of the dynamics of a situation, the recognition of people or objects, and collection of enough parallax information to accomplish these tasks requires a method of collecting multiple parallax information.
Collecting this type of information may be accomplished via biomimetic vision systems like a compound eye. Current methods of fabricating devices which simulate compound eyes are either lacking in functionality or are too small to interface. An exemplary device fabricated by Dr Luke Lee (see K. Jeong, J. Kim, L. P. Lee, Polymeric Synthesis of Biomimetic Artificial Compound Eyes, Proc. Transducers, Seoul, Korea, pp. 1110-1114, Jun. 5-9, 2005, Inspirations from Biological Optics for Advanced Photonic Systems Luke P. Lee and Robert Szema, 18 Nov. 2005 VOL 310 SCIENCE, Biologically Inspired Artificial Compound Eyes, Ki-Hun Jeong, Jaeyoun Kim, Luke P. Lee, SCIENCE VOL 312 28 Apr. 2006) does not have foveae and therefore misses an important feature of compound eyes. A replication technique employed through a molding process by the University of North Carolina, as reported by Kevin J. Henderson under NASA URETI “Biologically Inspired Materials” Grant no. NAG-1-23-1 Poster report 12/2005, is difficult to effectively interface based on its very small size.
It is therefore desirable to produce variable size compound eyes with foveae to provide the desired multiple parallax vision capability.
It is further desirable to provide a process for production of such lens systems with high repeatability and non-complex processing.